Qingyong

Hi @WenshuangSong, thanks for your interest in our work! Of course not. You'd better modify the code based on your dataset.

Hi @artless-spirit, Thanks for your interest in our work! Please refer to the visualization code [here](https://github.com/QingyongHu/RandLA-Net/blob/3926be6658221172b680d38d475551fda6d589b5/utils/6_fold_cv.py#L33-L38) Best,

Thanks for your interest in our work! @makaay2077 @ClosedOn It seems the data is not loaded correctly, please check if the data preparation is correct, and the path to the...

Hi @Asher-1， many thanks for your interest in our work. We haven't tested this code in windows, so can you please try this on the recommended configurations?

Thanks for pointing out this issue! Yes, we incorrectly calculate the class weights from the full annotations, which actually should calculate based on the sparse annotations. Actually, the class weights...

Sorry for my late response, the sharing code is `viso`. In addition, we also provide the Google drive download link, please check the GitHub page for details.

@csdingithub Thanks for your interest in our work! We have updated the dataset in Google drive and Baidu Yun, please refer to the new download link for the MOT dataset....

Hi @hzxie, thanks for your interest and valuable suggestions for our work. This is an ongoing work and we will add this to the next version of the preprint.

Hi @jongchyisu, thanks for your interest in our work. We will update the project page and preprint in the next version.

Thank you for bringing this to my attention. Can you please try to run more epochs on the model to see if the results improve?